Stories
Slash Boxes
Comments

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

An Overview of Parallelism

Posted by kdawson on Sat Feb 10, 2007 07:55 PM
from the cores-galore dept.
Mortimer.CA writes with a recently released report from Berkeley entitled "The Landscape of Parallel Computing Research: A View from Berkeley: "Generally they conclude that the 'evolutionary approach to parallel hardware and software may work from 2- or 8-processor systems, but is likely to face diminishing returns as 16 and 32 processor systems are realized, just as returns fell with greater instruction-level parallelism.' This assumes things stay 'evolutionary' and that programming stays more or less how it has done in previous years (though languages like Erlang can probably help to change this)." Read on for Mortimer.CA's summary from the paper of some "conventional wisdoms" and their replacements.

Old and new conventional wisdoms:
  • Old CW: Power is free, but transistors are expensive.
  • New CW is the "Power wall": Power is expensive, but transistors are "free." That is, we can put more transistors on a chip than we have the power to turn on.

  • Old CW: Monolithic uniprocessors in silicon are reliable internally, with errors occurring only at the pins.
  • New CW: As chips drop below 65-nm feature sizes, they will have high soft and hard error rates.

  • Old CW: Multiply is slow, but load and store is fast.
  • New CW is the "Memory wall" [Wulf and McKee 1995]: Load and store is slow, but multiply is fast.

  • Old CW: Don't bother parallelizing your application, as you can just wait a little while and run it on a much faster sequential computer.
  • New CW: It will be a very long wait for a faster sequential computer (see above).
+ -
story
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More
Loading... please wait.
  • by macadamia_harold (947445) on Saturday February 10 2007, @07:57PM (#17967086) Homepage
    Mortimer.CA writes with a recently released report from Berkley entitled "The Landscape of Parallel Computing Research: A View from Berkeley

    Would that be a Parallelograph?
  • nothing new here... (Score:3, Informative)

    by Anonymous Coward on Saturday February 10 2007, @08:08PM (#17967166)
    pretty much the same thing Dave Patterson's been saying for a while now...in fact, the CW sounded so familiar, I went back to double check his lecture slides from more than a year ago:

    http://vlsi.cs.berkeley.edu/cs252-s06/images/1/1b/ Cs252s06-lec01-intro.pdf [berkeley.edu]

    and it's pretty much identical (check out slide 3 on the first page of the pdf)
  • Erlang (Score:5, Insightful)

    by Anonymous Coward on Saturday February 10 2007, @08:09PM (#17967176)
    Erlang only provides a way of proving parallel correctness, a la CSP. This means avoiding deadlocks and such. The primary difficulty of crafting algorithms to run efficently over multiple CPUs still remains. Erlang does not do any automatic parallelization, and expects the programmer to write the code with multiple CPUs in mind.


    I'm wating for a language which would parallelize stuff for you. This is most likely to be a functinal language, or an extension to an existing functional language. Maybe even Erlang.

    • Re: (Score:3, Insightful)

      I wrote some parellel code using MPI [lam-mpi.org] in university. It takes a lot of work to get the hang of at first, and many people who I know that were good at programming had lots of trouble in this course, because programming for parallelism is very different than programming for a single processor. On the other hand, you can get much better performance from parallel algorithms. However, I think that we could do just as well sticking with the regular algorithms, and having a lot of threads each running on a diffe
    • Re:Erlang (Score:5, Informative)

      by cpm80 (899906) on Sunday February 11 2007, @12:19AM (#17968946) Homepage
      I think using a *specific* language for automatic parallelization is the wrong way. Some GNU folks are working on language independent autoparallelization for GCC 4.3. Their implementation is an extension to the OpenMP implementation in GCC. Read OpenMP and automatic parallelization in GCC, D. Novillo, GCC Developers' Summit, Ottawa, Canada, June 2006 http://people.redhat.com/dnovillo/Papers/ [redhat.com] for details.
      • Erlang also has a huge amount of overhead, and due to immutable data structures, has to spend a lot of time copying data around.

        This is the problem inherent with pure languages. Compilers/runtime systems are simply not sophisticated enough yet to reason about and perform as well as a human can with mutable data structures.

        Haskell comes pretty close, and it's designed from the beginning to be pure.

        In fact, it may be these immutable data structures that make the pure functional languages able to perform well

  • It's not hard (Score:5, Insightful)

    by PhrostyMcByte (589271) <phrosty@gmail.com> on Saturday February 10 2007, @08:12PM (#17967204) Homepage

    I think the main reason people say "don't use threads" is because while single threaded apps are easy to debug, multi-threaded ones will crash and burn at seemingly random places if the programmer didn't plan ahead and use proper locking. This is probably good advice to a noob programmer but I otherwise can't stand people who are of the "absolutely, never, ever, use threads" mindset.

    Some applications have no need to be multithreaded, but when they do it is a lot easier than people make it out to be. Taking advantage of lock-free algorithms and NUMA for maximum scalability *can* be hard, but the people who need these will have the proper experience to tackle it.

    Language extensions for threading would be great, and I'm sure somebody is working on it. But until that magical threading language (maybe c++1x) comes along the current ones work just fine.

    • Re:It's not hard (Score:4, Interesting)

      by ardor (673957) on Saturday February 10 2007, @08:23PM (#17967302)
      Well the indeterministic nature of multithreading is still a problem. With one thread, debugging is simple: the only thread present will be stopped. But with multiple threads, how is the debugger supposed to handle the problem? Stop all threads? Only the current one? Etc. This is relevant when debugging race conditions.

      Also, the second great problem is that thread problems are hard to find. When I write a class hierarchy, an OOP language can help me with seeing design errors (for example, unnecessary multiple inheritance), or misses in const-correctness. Threading, however, is only present as mutexes, conditions etc.

      One other issue with threads is that they effectively modify the execution sequence. Traditional single-threaded programs have a sequence that looks like a long line. Threading introduces branches and joins, turning the simple line into a net. Obviously, this complicates things. Petri nets can be useful in modeling this.
          • Re:It's not hard (Score:4, Insightful)

            by mikael (484) on Saturday February 10 2007, @10:07PM (#17967998)
            Not all algorithms can be parallelized that easily. Imagine e.g. a parser: You cannot parse text by having a million processors looking at one character each.

            You could have the first thread processor split the text by white space. Then each block of characters is assigned to any number of processors to find the matching token. I've seen some parsers where the entire document was
            read in, converted into an array of tokens before returning back to the calling routine.
    • by jd (1658) <.moc.oohay. .ta. .kapimi.> on Saturday February 10 2007, @09:54PM (#17967912) Homepage Journal
      This problem was "solved" (on paper) in the mid 1970s. Instead of writing a highly complex parallel program that you can't readily debug, you write a program that the computer can generate the parallel code for. Provided the compiler is correct, the sequential source and the parallel binary will be functionally the same, even though (at the instruction level) they might actually be quite different. What's more, if you compile the sequential source into a sequential binary, the sequential binary will behave exactly the same as the parallel version (only much slower).

      Any reproducable bug in the parallel binary will be reproducable given the same set of inputs on the sequential binary, which you can then debug as you have the corresponding sequential source code.

      So why isn't this done? Automagically parallelizing compilers (as opposed to compilers that merely parallelize what you tell them to parallelize) are extremely hard to write. Until the advent of Beowulf clusters, low-cost SMP and low-cost multi-core CPUs, there simply haven't been enough machines out there capable of sufficiently complex parallelism to make it worth the cost. Simply make a complex-enough inter-process communication system, with a million ways to signal and a billion types of events. Any programmer who complains they can't use that mess can then be burned at the stake for their obvious lack of appreciation for all these fine tools.

      Have you ever run GCC with maximum profiling over a program, tested the program, then re-run GCC using the profiling output as input to the optimizer? It's painful. Now, to parallelize, the compiler must automatically not just do one trivial run but get as much coverage as possible, and then not just tweak some optimizer flags but run some fairly hefty herustics to guess what a parallel form might look like. And it will need to do this not just the once, but many times over to find a form that is faster than the sequential version and does not result in any timing bugs that can be picked up by automatic tools.

      The idea of spending a small fortune on building a compiler that can actually do all that reliably, effectively, portably and quickly, when the total number of purchasers will be in the double or treble digits at most - say what you like about the blatant stupidity rife in commercial software, but they know a bad bet when they see one. You will never see something with that degree of intelligence come out of PCG or Green Hills - if they didn't go bankrupt making it, they'd go bankrupt from the unsold stock, and they know it.

      What about a free/open source version? GCC already has some of the key ingredients needed, after all. Aside from the fact that the GCC developers are not known for their speed or responsiveness - particularly to arcane problems - it would take many days to compile even SuperTuxKart and probably months when it came to X11, glibc or even the Linux kernel. This is far longer than the lifetime of most of the source packages - they've usually been patched on that sort of timeframe at least once. The resulting binaries might even be truly perfectly parallel, but they'd still be obsolete. You'd have to do some very heavy research into compiler theory to get GCC fast enough and powerful enough to tackle such problems within the lifetime of the product being compiled. Hey, I'm not saying GCC is bad - as a sequential, single-pass compiler, it's pretty damn good. At the Supercomputer shows, GCC is used as the benchmark to beat, in terms of code produced. The people at such shows aren't easily impressed and would not take boasts of producing binaries a few percent faster than GCC unless that meant a hell of a lot. But I'm not convinced it'll be the launchpad for a new generation of automatic parallelizing compilers. I think that's going to require someone writing such a compiler from scratch.

      Automatic parallelization is unlikely to happen in my lifetime, even though the early research was taking place at about the time I first started primary school. It's a hard problem that isn't being made easier by having been largely avoided.

  • Hmmm... (Score:5, Interesting)

    by ardor (673957) on Saturday February 10 2007, @08:13PM (#17967216)
    "but is likely to face diminishing returns as 16 and 32 processor systems are realized"

    Then we are doing something wrong. The human brain provides compelling evidence that massive parallelization works. So: what are we missing?
    • Re:Hmmm... (Score:5, Funny)

      by CosmeticLobotamy (155360) on Saturday February 10 2007, @08:32PM (#17967362)
      Then we are doing something wrong. The human brain provides compelling evidence that massive parallelization works. So: what are we missing?

      Brain scalability is just not that great. Trust me, putting more than four brains in one head is just asking for locking problems out the yin-yang.
    • Re: (Score:3, Interesting)

      Is this really true? Of course for some tasks the massive parallelism of the human brain works great. The brain can analyze complex images extremely fast, comparing them in parallel to it's own internal database of images, using fuzzy reasoning to detect if something is familiar etc. However you give your brain complex math problems, and it can spend seconds, minutes, or even hours to solve it, sometimes requiring extra scratch memory to solve. This is due to bad programming in the brain that sucks at d
  • by SparhawkA (608965) on Saturday February 10 2007, @08:16PM (#17967236)
    Take a look at LabVIEW, a compiled graphical programming language from National Instruments. It natively supports SMP / multicore / multithreading. Essentially, dissociated pieces of code you write (computations, hardware I/O, etc.) are automatically scheduled in separate threads of execution in order to maximize efficiency. It's an interesting idea: here's a technical article from their website that does a better job of describing it (some marketing included as well): http://zone.ni.com/devzone/cda/tut/p/id/4233 [ni.com]
  • From TFA (Score:5, Funny)

    by obender (546976) on Saturday February 10 2007, @08:17PM (#17967240)

    The target should be 1000s of cores per chip
    640 cores should be enough for anyone.
  • by RAMMS+EIN (578166) on Saturday February 10 2007, @08:25PM (#17967312) Homepage Journal
    I think parallelism can be achieved elegantly using languages that express what is to be done, rather than how it is to be done. Functional programming is a major step in the right direction. Not only do functional programs typically more clearly express what is to be done (as opposed to which steps are to be taken to get there), they also tend to cause fewer side effects (which restrict the correct evaluation orders). In particular, not using variables avoids many of the headaches traditionally involved in multithreading.
    • Re: (Score:3, Insightful)

      Functional languages are no silver bullet, however. Things like I/O do not fit well in there. Yes, there are solutions for this, but they tend to be overly complicated. A hybrid functional/imperative language with safeguards for side-effects of the imperative parts seems to be the way to go.
  • by Animats (122034) on Saturday February 10 2007, @08:59PM (#17967546) Homepage

    I just heard that talk; he gave it at EE380 at Stanford a few weeks ago.

    First, this is a supercomputer guy talking. He's talking about number-crunching. His "13 dwarfs" are mostly number-crunching inner loops. Second, what he's really pushing is getting everybody in academia to do research his way - on FPGA-based rackmount emulators.

    Basic truth about supercomputers - the commercial market is zilch. You have to go down to #60 on the list of the top 500 supercomputer [top500.org] before you find the first real commercial customer. It's BMW, and the system is a cluster of 1024 Intel x86 1U servers, running Red Hat Linux. Nothing exotic; just a big server farm set up for computation.

    More CPUs will help in server farms, but there we're I/O bound to the outside world, not talking much to neighboring CPUs. If you have hundreds of CPUs on a chip, how do you get data in and out? But we know the answer to that - put 100Gb/s Ethernet controllers on the chip. No major software changes needed.

    This brings up one of the other major architectural truths: shared memory multiprocessors are useful, and clusters are useful. Everything in between is a huge pain. Supercomputer guys fuss endlessly over elaborate interconnection schemes, but none of them are worth the trouble. The author of this paper thinks that all the programming headaches of supercomputers will have to be brought down to desktop level, but that's probably not going to happen. What problem would it solve?

    What we do get from the latest rounds of shrinkage are better mobile devices. The big wins commercially are in phones, not desktops or laptops. Desktops have been mostly empty space inside for years now. In fact, that's true of most non-mobile consumer electronics. We're getting lower cost and smaller size, rather than more power.

    Consider cars. For the first half of the 20th century, the big thing was making engines more powerful. By the 1960s, engine power was a solved problem, (the 1967 turbine-powered Indy car finally settled that issue) and cars really haven't become significantly more powerful since then. (Brakes and suspensions, though, are far better.)

    It will be very interesting to see what happens with the Cell. That's the first non-shared memory multiprocessor to be produced in volume. If it turns out to be a dead end, like the Itanium, it may kill off interest in that sort of thing for years.

    There are some interesting potential applications for massive parallelism for vision and robotics applications. I expect to see interesting work in that direction. The more successful vision algorithms do much computation, most of which is discarded. That's a proper application for many-CPU machines, though not the Cell, unless it gets more memory per CPU. Tomorrow's robots may have a thousand CPUs. Tomorrow's laptops, probably not.

    • Re: (Score:3, Interesting)

      Basic truth about supercomputers - the commercial market is zilch. You have to go down to #60 on the list of the top 500 supercomputer before you find the first real commercial customer.

      You may want to adjust your truth as your measure of the market is wrong. The Top500 is not a marketing survey and just because you have HPC hardware does mean you run out and try an get it on the Top500. Many companies are using (HPC) parallel cluster computers, but they choose to be quiet about it for competitive reason

      • Re: (Score:3, Insightful)

        Also keep in mind that many companies aren't interested in linpac peformance per se, at least to the extent that they will spend a lot of time and effort tweaking their computers to get really high linpac scores, which is all that is important when it comes to top500.
    • by RecessionCone (1062552) on Saturday February 10 2007, @10:05PM (#17967986)
      I don't think you were listening very carefully to the talk (or know much about Computer Architecture) if you think Dave Patterson is a supercomputer guy. Perhaps you've heard of the Hennessy & Patterson Quantitative Approach to Computer Architecture book (you know, the one used at basically every university to teach about computer architecture). Patterson has been involved in a lot of different things within computer architecture over the years, including being one of the main people behind RISC and RAID (as well as being the president of the ACM). I saw his talk when it was given at Berkeley, and you really missed the point if you thought it was about supercomputing. The talk was about the future of computing in general, which is increasingly parallel, in case you're unaware of that fact. GPUs are already at 128 cores, Network processors are up to 200 cores. Intel is going to present an 80 core x86 test chip tomorrow at ISSCC. Physics won't support faster single core processors at the rate we're accustomed to, so the whole industry is going parallel, which is a sea change in the industry. Patterson's talk is aimed at the research community, since we don't have good answers as to how these very parallel systems should be architected and programmed. FPGA emulation is a great way to play around with massive multiprocessor configurations and programming strategies, which is why Patterson is advocating it (his RAMP project has researchers from MIT, Berkeley, Stanford, Texas, Washington involved (among others)). You also need to have a little more imagination about what we could do with more computing power. Try looking at Intel's presentations on RMS http://www.intel.com/technology/itj/2005/volume09i ssue02/foreword.htm [intel.com].
  • by deadline (14171) on Saturday February 10 2007, @09:14PM (#17967630) Homepage

    Those of us that use HPC clusters (i.e. Beowulf) have been thinking about these issues as well. For those interested, I wrote a series of articles on how one might program 10,000 cores (based on my frustrations as programmer and user of parallel computers). Things will change, there is no doubt.

    The first in the series is called Cluster Programming: You Can't Always Get What You Want [clustermonkey.net] The next two are Cluster Programming: The Ignorance is Bliss Approach [clustermonkey.net], and Cluster Programming: Explicit Implications of Cluster Computing [clustermonkey.net].

    Comments welcome.

  • by zestyping (928433) on Saturday February 10 2007, @09:59PM (#17967950) Homepage
    Reliably achieving even simple goals using concurrent threads that share state is extremely difficult. For example, try this task:

    Implement the Observer [wikipedia.org] (aka Listener) pattern (specifically the thing called "Subject" on the Wikipedia page). Your object should provide two methods, publish and subscribe. Clients can call subscribe to indicate their interest in being notified. When a client calls publish with a value, your object should pass on that value by calling the notify method on everyone who has previously subscribed for updates.

    Sounds simple, right? But wait:
    • What if one of your subscribers throws an exception? That should not prevent other subscribers from being notified.
    • What if notifying a subscriber triggers another value to be published? All the subscribers must be kept up to date on the latest published value.
    • What if notifying a subscriber triggers another subscription? Whether or not the newly added subscriber receives this in-progress notification is up to you, but it must be well defined and predictable.
    • Oh, and by the way, don't deadlock.
    Can you achieve all these things in a multithreaded programming model (e.g. Java)? Try it. Don't feel bad if you can't; it's fiendishly complicated to get right, and i doubt i could do it.

    Or, download this paper [erights.org] and start reading from section 3, "The Sequential StatusHolder."

    Once you see how hard it is to do something this simple, now think about the complexity of what people regularly try to achieve in multithreaded systems, and that pretty much explains why computer programs freeze up so often.
    • by master_p (608214) on Sunday February 11 2007, @07:45AM (#17971256)
      There is a way to automate shared state concurrency! every object should be its own thread. Computations that refer to the same object must be executed by the object's thread.

      Here is how it works:

      A computation does not return a result, but a tuple of {key, continuation}. The key is used to locate the thread to pass the continuation to. The computation is stored in the thread's queue and the thread is woken up.

      The tuple {key, continuation} pair can be an 64-bit value (on 32-bit machines) that consists of a pointer to a memory location (the key) and a pointer to code (the continuation).

      The insertion to the thread's queue can be done using lock-free data structures.

      Threads can be user-level so there need not be a switch to kernel space.

      This design can allow for linear scaling of performance: the more cores you put in, the more performance you get (for algorithms that are not linear, that is). Linear algorithms would execute a little slower than usual, but the trade off is acceptable: for many applications that allow for parallelization due to having lots of (relatively) independent objects, the performance boost be tremendous.

      There are many domains of applications that would benefit from such an approach:

      -web servers/application servers that must serve thousands of clients simultaneously.
      -video games with thousands of objects.
      -simulations that have many independent agents that can run in parallel.
      -GUI apps that use the observer pattern and each observable has many observers than can be notified in parallel.

      Note: The above ideas are taken from libasync-mp and lock-free data structure programming.

  • I used to work for SilverStorm (recently purchased by QLogic). They make InfiniBand switches and software for use in high performance computing and enterprise database clustering. The quality of the I/O subsystem of a cluster played a large part in determining the performance of a cluster. Latency (down the microsecond) and bandwidth (over 10 gigabits per second) both mattered.

    Also, we found that sometimes, what made a deal go through was how well your proposed system could run some prexisting software. For example, vendors would publish how well they could run a standard crash test simulation.

    Also, I would like to see more research put into making clustered operating systems like mosix good enough so that developers can stick to what they have learned on traditional SMP systems and have their code just work on large clusters. I don't think that multicore processors eliminate the need for better cluster software.